Gemini 4 Neon vs Muse Ember
TryTop 2 now
Gemini 4 Neon leads by 6.8 capability points (≈95% interval +4.0 to +9.6): a clear lead over Muse Ember · P(higher) ≈ 100%
Wins 47 of 49 shared benchmarks (2 lost)
Capability vs. relative cost
Gemini 4 NeonMuse EmberClosed weightsOpen weights
Benchmark wins47 – 2 · 49 shared
Comparing matching setups only (2 benchmarks excluded) · suites counted once
Area← Gemini 4 NeonMuse Ember →
Data: Artificial Analysis, Vals AI, Surge AI and 3 othersatlas.village.dev
Margins
Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Muse Ember minus Gemini 4 Neon.
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 70.0 | 17.6 | −52.4 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 71.3 | 25.1 | −46.2 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 18.4 | 1.0 | −34.0 |
| RiemannBench (November 2026 rescoring) | Math | 84.0 | 53.6 | −30.4 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 47.2 | 17.2 | −30.0 |
| Code Migration v1.1 (Vals Index v2.2 subset) | Software › Software engineering | 75.8 | 47.1 | −28.7 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 63.6 | 35.4 | −28.3 |
| GDP.xlsx | Professional › General knowledge work | 43.1 | 15.4 | −27.7 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 50.5 | 24.2 | −26.3 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 61.4 | 36.0 | −25.4 |
| Artificial Analysis Legal Index | Professional › Legal | 63.8 | 38.5 | −25.3 |
| Artificial Analysis Engineering Index | Science › Engineering | 66.8 | 41.8 | −25.1 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 44.5 | -2.7 | −23.6 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 54.8 | 7.7 | −23.5 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 52.3 | 29.8 | −22.5 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 44.4 | 22.0 | −22.4 |
| Terminal-Bench 4.0 | Software › Software engineering | 61.6 | 39.4 | −22.2 |
| SimpleBench | Knowledge & reasoning › General knowledge | 94.4 | 72.7 | −21.7 |
| Artificial Analysis Economics Index | Composite indexes | 61.9 | 41.0 | −21.0 |
| LMArena Code Arena WebDev | Software › Frontend & design | 1764 | 1608 | −20.7 |
| Artificial Analysis Finance & Accounting Index | Professional › Finance, accounting & tax | 65.1 | 44.6 | −20.5 |
| Artificial Analysis Coding Agent Index v2.1 | Software › Software engineering | 58.6 | 39.7 | −19.0 |
| Artificial Analysis Healthcare & Medical Index | Health & medicine | 60.1 | 41.3 | −18.8 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 77.4 | 62.0 | −15.4 |
| EnterpriseBench: CoreCraft | Tool & computer use › Customer service & operations | 77.4 | 62.1 | −15.3 |
| Artificial Analysis Strategy & Ops Index | Professional › General knowledge work | 63.8 | 48.7 | −15.1 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 72.9 | 58.8 | −14.1 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 31.7 | 18.3 | −13.3 |
| APEX-Agents 1.1 | Professional › General knowledge work | 88.9 | 75.7 | −13.2 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 79.3 | 68.1 | −11.3 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 33.4 | 22.8 | −10.6 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 19.2 | -2.0 | −10.6 |
| Vals Index v2.2 | Composite indexes | 70.9 | 60.5 | −10.4 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 93.7 | 83.7 | −10.0 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1699 | 1526 | −9.5 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 79.8 | 71.4 | −8.4 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 21.2 | 4.8 | −8.2 |
| Vals Finance Index | Professional › Finance, accounting & tax | 68.3 | 60.9 | −7.4 |
| Chartography (November 2026 rescoring) | Perception & media › Documents & charts | 86.6 | 79.3 | −7.3 |
| Legal Research Bench | Professional › Legal | 56.3 | 49.0 | −7.2 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | -6.0 | 6.6 | +6.3 |
| τ³-Banking (Mercor run, bm25_grep retrieval) | Tool & computer use › Customer service & operations | 59.5 | 53.3 | −6.2 |
| HLAB-Vals-1.1 | Professional › Legal | 19.6 | 25.0 | +5.4 |
| Finance Agent v2.1 | Professional › Finance, accounting & tax | 68.6 | 63.8 | −4.7 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 76.5 | 72.2 | −4.3 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 53.4 | 49.2 | −4.2 |
| Harvey LAB (Mercor run, undisclosed ~213-task cohort) | Professional › Legal | 26.6 | 22.7 | −3.9 |
| Vals Legal Index | Professional › Legal | 78.4 | 75.7 | −2.7 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 1.0 | 0.9 | 0.0 |
Data: Artificial Analysis, Vals AI, Surge AI and 3 othersatlas.village.dev